def grimm_filter_unique_gene(in_file, out_file):
lines = open(in_file).read().split('\n')
# make unique blocks list
i = 0
flt = Unique_Filter()
while i < len(lines):
line = lines[i]
if GRIMMReader.is_genome_declaration_string(line):
data_line = lines[i + 1]
parsed = GRIMMReader.parse_data_string(data_line)[1]
flt.update_allowed_blocks(parsed)
i += 2
else:
i += 1
# write allowed blocks
i = 0
with open(out_file, 'w') as f:
while i < len(lines):
line = lines[i]
if GRIMMReader.is_genome_declaration_string(line):
data_line = lines[i + 1]
parsed = GRIMMReader.parse_data_string(data_line)[1]
parsed = flt.filter_unique(parsed)
print(line, file=f)
print(' '.join(p[0] + p[1] for p in parsed), '@', file=f)
i += 2
else:
i += 1
return list(map(int, flt.allowed_blocks))
def test_parse_data_string_correct(self):
# data string is parsed by getting information about genes order and individual orientations for each block (gene)
# string based processing if performed
# if no orientation is specified explicitly, positive orientation is assumed
data_string = "a $"
result = GRIMMReader.parse_data_string(data_string)
self.assertEqual(result[0], "$")
self.assertEqual(result[1][0][0], "+")
self.assertEqual(result[1][0][1], "a")
self.assertEqual(len(result[0]), 1)
self.assertEqual(len(result[1]), 1)
data_string = "a @"
result = GRIMMReader.parse_data_string(data_string)
self.assertEqual(result[0], "@")
self.assertEqual(result[1][0][0], "+")
self.assertEqual(result[1][0][1], "a")
self.assertEqual(len(result[0]), 1)
self.assertEqual(len(result[1]), 1)
data_string = " a -b c -d @ e f "
result = GRIMMReader.parse_data_string(data_string)
self.assertEqual(result[0], "@")
reference_genes = ["a", "b", "c", "d"]
result_genes = [gene[1] for gene in result[1]]
reference_signs = ["+", "-", "+", "-"]
result_signs = [gene[0] for gene in result[1]]
self.assertListEqual(result_genes, reference_genes)
self.assertListEqual(result_signs, reference_signs)
data_string = " a -b +c -d $ e f "
result = GRIMMReader.parse_data_string(data_string)
self.assertEqual(result[0], "$")
reference_genes = ["a", "b", "c", "d"]
result_genes = [gene[1] for gene in result[1]]
reference_signs = ["+", "-", "+", "-"]
result_signs = [gene[0] for gene in result[1]]
self.assertListEqual(result_genes, reference_genes)
self.assertListEqual(result_signs, reference_signs)
data_string = " a -b c -d @ e f $ g -h "
result = GRIMMReader.parse_data_string(data_string)
self.assertEqual(result[0], "@")
reference_genes = ["a", "b", "c", "d"]
result_genes = [gene[1] for gene in result[1]]
reference_signs = ["+", "-", "+", "-"]
result_signs = [gene[0] for gene in result[1]]
self.assertListEqual(result_genes, reference_genes)
self.assertListEqual(result_signs, reference_signs)
def test_parse_data_string_error(self):
# data string must contain a fragment termination symbol ($ or @)
# and must contain space separated gene order information before fragment termination symbol
data_string_1 = " a b c d e "
data_string_2 = ""
data_string_3 = " a -b -c d -e "
data_string_4 = "$"
data_string_5 = "@"
data_string_6 = "@ a d s d"
data_string_7 = "$a d s d"
data_string_8 = "$-a d s d"
data_string_9 = "@+a d s d"
data_string_10 = "a b - -c d e $"
for data_string in [
data_string_1, data_string_2, data_string_3, data_string_4,
data_string_5, data_string_6, data_string_7, data_string_8,
data_string_9, data_string_10
]:
with self.assertRaises(ValueError):
GRIMMReader.parse_data_string(data_string)
def get_block_neighbours(grimm_file):
block_neighbours = defaultdict(lambda: defaultdict(list))
with open(grimm_file) as f:
ls = f.readlines()
i = 0
while i < len(ls):
l = ls[i]
if GRIMMReader.is_genome_declaration_string(l):
genome = GRIMMReader.parse_genome_declaration_string(l)
data_line = ls[i + 1]
bs = GRIMMReader.parse_data_string(data_line)[1]
n = len(bs)
j = 0
while j < n:
tandem_copies = 1
prev_or, prev_block = bs[j % n]
_, curr_block = bs[(j + 1) % n]
next_or, next_block = bs[(j + 2) % n]
if curr_block == prev_block:
j += 1
continue
while curr_block == next_block:
j += 1
tandem_copies += 1
next_or, next_block = bs[(j + 2) % n]
neighbours = (prev_block + ('h' if prev_or == '+' else 't'),
next_block + ('t' if next_or == '+' else 'h'))
orientations = tuple(bs[(k + 1) % n][0]
for k in range(j - tandem_copies + 1, j +
1))
if orientations[0] == '-':
neighbours = (neighbours[1], neighbours[0])
orientations = tuple('+' if or_ == '-' else '+'
for or_ in orientations[::-1])
block_neighbours[int(curr_block)][genome.name].append(
(*neighbours, tandem_copies, orientations))
j += 1
i += 2
else:
i += 1
return block_neighbours
def get_genomes_contain_blocks_grimm(grimm_file):
genomes, blocks = set(), set()
with open(grimm_file) as f:
ls = f.readlines()
block_genome_count = defaultdict(Counter)
for i in range(0, len(ls), 2):
name = GRIMMReader.parse_genome_declaration_string(ls[i]).name
data = GRIMMReader.parse_data_string(ls[i + 1])[1]
genomes.add(name)
for _, block in data:
blocks.add(int(block))
block_genome_count[int(block)][name] += 1
return list(sorted(genomes)), list(sorted(blocks)), block_genome_count
with open(file_name, "rt") as source:
current_genome = None
for line in source:
line = line.strip()
if len(line) == 0 or GRIMMReader.is_comment_string(
data_string=line):
continue
if GRIMMReader.is_genome_declaration_string(data_string=line):
current_genome = GRIMMReader.parse_genome_declaration_string(
data_string=line).name
if args.trim_names:
current_genome = current_genome.split(
args.trimmer_char, 1)[0]
elif current_genome is not None:
current_chromosome = []
chr_type, blocks = GRIMMReader.parse_data_string(
data_string=line)
genomes[current_genome].append((chr_type, blocks))
if args.good_genomes != "":
good_genomes = args.good_genomes.split(",")
if args.trim_names:
good_genomes = [
genome_name.split(args.trimmer_char, 1)[0]
for genome_name in good_genomes
]
for genome_name in list(genomes.keys()):
if genome_name not in good_genomes:
del genomes[genome_name]
if args.bad_genomes != "":
bad_genomes = args.bad_genomes.split(",")
if args.trim_names:
bad_genomes = [
genomes = defaultdict(list)
for file_name in args.grimm:
logger.info("Processing file \"{file_name}\"".format(file_name=file_name))
with open(file_name, "rt") as source:
current_genome = None
for line in source:
line = line.strip()
if len(line) == 0 or GRIMMReader.is_comment_string(data_string=line):
continue
if GRIMMReader.is_genome_declaration_string(data_string=line):
current_genome = GRIMMReader.parse_genome_declaration_string(data_string=line).name
if args.trim_names:
current_genome = current_genome.split(args.trimmer_char, 1)[0]
elif current_genome is not None:
current_chromosome = []
chr_type, blocks = GRIMMReader.parse_data_string(data_string=line)
genomes[current_genome].append((chr_type, blocks))
if args.good_genomes != "":
good_genomes = args.good_genomes.split(",")
if args.trim_names:
good_genomes = [genome_name.split(args.trimmer_char, 1)[0] for genome_name in good_genomes]
for genome_name in list(genomes.keys()):
if genome_name not in good_genomes:
del genomes[genome_name]
if args.bad_genomes != "":
bad_genomes = args.bad_genomes.split(",")
if args.trim_names:
bad_genomes = [genome_name.split(args.trimmer_char, 1)[0] for genome_name in bad_genomes]
for genome_name in list(genomes.keys()):
if genome_name in bad_genomes:
del genomes[genome_name]